US7143054B2 - Assessment of communication strengths of individuals from electronic messages - Google Patents
Assessment of communication strengths of individuals from electronic messages Download PDFInfo
- Publication number
- US7143054B2 US7143054B2 US09/897,577 US89757701A US7143054B2 US 7143054 B2 US7143054 B2 US 7143054B2 US 89757701 A US89757701 A US 89757701A US 7143054 B2 US7143054 B2 US 7143054B2
- Authority
- US
- United States
- Prior art keywords
- individuals
- individual
- strengths
- communication
- talker
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime, expires
Links
- 238000004891 communication Methods 0.000 title claims abstract description 74
- 230000000694 effects Effects 0.000 claims abstract description 50
- 238000000034 method Methods 0.000 claims abstract description 37
- 230000003612 virological effect Effects 0.000 claims abstract description 33
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 238000011160 research Methods 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 31
- 238000010586 diagram Methods 0.000 claims description 30
- 238000003306 harvesting Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 11
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 238000004445 quantitative analysis Methods 0.000 description 24
- 230000015654 memory Effects 0.000 description 11
- 238000007418 data mining Methods 0.000 description 7
- 238000013515 script Methods 0.000 description 6
- 230000009977 dual effect Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 239000000047 product Substances 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000003467 diminishing effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000003466 anti-cipated effect Effects 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 238000000746 purification Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 241000282326 Felis catus Species 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000004883 computer application Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000035622 drinking Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008450 motivation Effects 0.000 description 1
- 230000007425 progressive decline Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003997 social interaction Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000000153 supplemental effect Effects 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0204—Market segmentation
Definitions
- the invention is generally related to marketing and computer analysis of electronic media in connection with the same. More particularly, the invention is related to the quantitative analysis of electronic messages for the purpose of improving marketing efficacy.
- Marketing techniques have substantially improved in terms of sophistication and efficacy since the days of undirected mass-market advertisements and the like.
- Marketing campaigns desirably balance the number of “impressions” or contacts with consumers with the quality of those impressions and the overall cost of the campaign.
- the efficacy of a marketing campaign is largely premised on getting the most “bang for the buck”, i.e., the greatest impact for the marketing dollar.
- viral marketing One specific type of marketing technique is known as viral, or diffusion, marketing.
- viral marketing the recipient of an encounter or impression is desirably used to relay a marketing message to other consumers.
- Viral marketing thus exemplifies a somewhat more sophisticated form of “word of mouth” advertising.
- To maximize the efficacy of a viral marketing campaign therefore, it is desirable to direct encounters toward only those recipients who are most likely to relay marketing messages to others, and thus who are most capable of maximizing the viral relay effect.
- Mavens are typically collectors of information who are motivated to distribute information to other individuals, predominantly due to a social motivation.
- Connectors are individuals who communicate information of interest to themselves to an unusually large number of other people. Some connectors may merely be “enthusiasts” who effectively relay information to others only pertaining to a narrow area of interest. Other connectors, however, may be likely to convey information to others on a wide number of topics outside of their particular areas of interest. For example, a connector who has a particular area of expertise in musical instruments might still be inclined to relate to others about a wonderful dining experience that he or she recently had at a particular restaurant.
- the Internet in particular, is interesting from a sociological standpoint for the manner in which “online communities” are formed by different individuals based upon particular areas of interest to those individuals.
- the very nature of the Internet and in particular the subset of the Internet known as the “World Wide Web” (or simply “the Web”), presents social and marketing researchers with a vast repository of data from which useful trends and relationships may be obtained. For example, research has been devoted to identifying the kinds and qualities of the groups and institutions that people create and use on the Internet, as well as to spot online trends and/or detect customer likes and dislikes from forums and other enthusiast communities, etc.
- One particular avenue of research is based on the premise that the authority of a particular content item can be assessed based upon the number of links to that item by other content items. Furthermore, to make such an assessment, content items are segregated into “hubs” and “authorities”. Hubs are typically directory-type content items that provide relatively large numbers of links to other content items, much like a telephone directory. Authorities are typically the end use content items that are desirably returned as search results. Assessment of authority is premised further on the dual propositions that (1) a good hub links to many good authorities, and (2) a good authority is linked to by many good hubs.
- search results may be arranged so as to favor content items that are assessed to have greater authority by virtue of their links to greater numbers and qualities of other content items.
- the Internet presents an ever-growing repository of information regarding individuals and the manners in which they communicate, in particular with respect to the electronic messages that individuals author and convey to others.
- the electronic messages created and maintained within electronic environments and media such as news archives and forums, list servers, chat rooms, etc. often present a wealth of information regarding the individuals that participate in such environments.
- the messages are relatively well categorized and capable of being searched to harvest relevant information about a particular topic.
- Internet forums are typically devoted to a particular area or field of interest. With those forums, members often post messages about particular topics or questions that interest them, and other members reply to those messages with answers or further discussions.
- the original messages and the replies thereto are often grouped into “threads” that present entire discussions in a cohesive manner.
- the invention addresses these and other problems associated with the prior art by providing an apparatus, program product, and method that analyze the various electronic messages directed between multiple individuals to quantitatively assess the relative communication capabilities of such individuals.
- an apparatus, program product, and method that analyze the various electronic messages directed between multiple individuals to quantitatively assess the relative communication capabilities of such individuals.
- a quantitative value referred to as “communication strength” is determined for one or more individuals from a plurality of individuals based upon analysis of the electronic messaging activities of such individual(s), and is subsequently used to select an individual from among the plurality of individuals, e.g., for the purpose of selecting a target for a viral marketing encounter, a market research project, etc.
- the communication strength of an individual is determined by modeling a plurality of conversations identified through analysis of a plurality of electronic messages generated by the plurality of individuals. Each conversation is associated with a talker and a listener from the plurality of individuals, and is useful in determining a level of messaging activity between the talker and listener associated with such conversation.
- FIG. 1 is a block diagram of a viral marketing architecture incorporating communication assessment of electronic media consistent with the invention.
- FIG. 2 is a block diagram of an apparatus suitable for implementing communication assessment in the viral marketing architecture of FIG. 1 .
- FIG. 3 is a flowchart illustrating the sequence of operations in performing communication assessment in the apparatus of FIG. 2 .
- FIG. 4 is a block diagram of an electronic medium archive, illustrating a first, restrictive talker-listener criterion.
- FIG. 5 is a block diagram of an electronic medium archive, illustrating a second, loose talker-listener criterion.
- FIGS. 6A and 6B are object diagrams illustrating a database schema for an exemplary conversation archive from the apparatus of FIG. 2 .
- FIG. 7 illustrates the program flow of an exemplary routine for implementing the build model step of FIG. 3 .
- FIG. 8 is a block diagram of an exemplary working matrix generated by the routine of FIG. 7 .
- FIG. 9 illustrates the program flow of an exemplary routine for implementing the solve step of FIG. 3 .
- FIG. 10 is an exemplary talker strength diagram capable of being generated by the routine of FIG. 9 .
- FIG. 11 is a block diagram of an exemplary study results display utilized in the report step of FIG. 3 .
- FIG. 12 is an exemplary bubble diagram capable of being generated by the routine of FIG. 9 .
- FIG. 13 is another exemplary talker strength diagram capable of being generated by the routine of FIG. 9 .
- communication assessment consistent with the invention may be utilized in connection with electronic messages of differing types, as well as electronic messages found in various electronic environments.
- communication assessment may be performed on various types of electronic messages such as emails, message posts, telephone records, “instant messages”, chat sessions, and combinations thereof, whether incorporating text, image data, animation, video, audio, and/or other electronic information.
- such messages may be retrieved from electronic environments such as news archives, forum archives, mailing (listserv) list archives, chat archives, instant messaging archives, telephone record archives, email archives, etc.
- a talker may be, for example, an author, sender and/or other originator of a message, while a listener is typically someone who is a recipient of a message.
- a listener is typically someone who is a recipient of a message.
- a study domain will include conversations where the “talker” for a particular conversation is an individual designated to be a “listener” for the study domain and/or where the “listener” for a particular conversation is an individual designated to be a “talker” for the study domain.
- conversations within a study domain may include talker-listener conversations, where the roles for the individuals within such conversations are consistent with those defined for the overall domain, as well as talker-talker, listener-talker, and listener-listener conversations, where one or both of the participants in such conversations have roles within such conversations that are inconsistent with their designated roles in the study domain.
- Precisely how an individual is designated as a talker or a listener can vary in different applications, and is discussed in greater detail below. It should also be appreciated that some electronic messages may be addressed to multiple recipients, whereby a particular message may be considered to represent multiple conversations between the same talker and different listeners. In the alternative, multiple listeners may be considered to be participating in the same conversation.
- Conversations are typically harvested and derived from selected messages that pertain to a particular selection criterion, e.g., to analyze only messages dealing with a particular area or domain of interest.
- the conversations identified during communication assessment may represent only a subset of the messages capable of being accessed by a communication assessment system consistent with the invention.
- Communication strength consistent with the invention may represent either or both of “talker” strength and “listener” strength, and may be dependent upon calculations made under differing conditions (e.g., using different wearout factors, as discussed in greater detail below).
- Talker strength generally represents an individual's propensity to communicate to others
- listener strength generally represents an individual's propensity to listen or otherwise be willing to receive information communicated by others.
- communication strength may also be representative of “connector” strength as well.
- identified conversations are used to determine the general levels of messaging activity between various talkers and listeners, typically in the form of activity indicators such as conversation counts. These activity indicators are then typically used to populate a matrix from which modeling may be performed to facilitate the quantitative assessment of communication strength.
- a quantitative assessment of communication strength as used herein is based upon the fundamental underlying propositions that: (1) strong talkers are those who frequently talk to strong listeners; and (2) strong listeners are those who frequently listen to strong talkers. As will become more apparent below, these two propositions may be resolved through solution of an eigenvalue problem using quantitative modeling techniques. From such modeling techniques, one or both of talker and listener strength may be derived for selected individuals, and under varying input conditions.
- the end uses of calculated communication strengths may include, for example, the selection of “connectors”, e.g., strong talkers (who are also possibly strong listeners), who are more likely to be able to effectively convey viral marketing messages, as well as have those messages listened to and given weight by other individuals, or who may be appropriate for use in market research projects.
- listener strength may also be utilized separate from talker strength to identify individuals that are particularly receptive to particular messages or encounters.
- FIG. 1 illustrates an exemplary application of communication assessment, in particular within a viral marketing architecture or environment 10 , in which is illustrated an exemplary set of individuals 12 .
- Individuals 12 may be considered to exist within either or both of a “cyber” world 14 (i.e., as those individuals interact with other individuals over the Internet or another electronic medium) or physical world 16 .
- an enterprise 18 may utilize a communication assessment system 20 to harvest information about electronic communications between individuals 12 , and from such information, to assess connection strengths of various individuals in connection with particular study parameters.
- the harvested information takes the form of electronic messages 22 .
- the harvested electronic messages are processed by communication assessment system 20 to generate communication strengths for various individuals so that one or more of such individuals may be selected from a pool of individuals based upon their respective communication strengths.
- communication strength e.g., having relatively high talker strengths, and optionally, relatively high listener strengths as well
- it may be desirable to select individuals having relatively communication strengths e.g., having relatively high talker strengths, and optionally, relatively high listener strengths as well
- connectors e.g., having relatively high talker strengths, and optionally, relatively high listener strengths as well
- FIG. 1 for example, individuals 12 that meet a connector criterion are illustrated using cross-hatched representations.
- an individual 24 within enterprise 18 e.g., a maven and/or salesperson, will attempt to initiate a viral marketing encounter with one or more connectors, e.g., as represented by encounter 25 directed to a connector individual designated at 12 A. Then, as shown in FIG. 1 , it is anticipated that the designated connector 12 A will relay the viral marketing message to other individuals 12 within either or both of the “cyber” and physical worlds 14 , 16 .
- Any number of known viral marketing encounters may be utilized to initiate a viral marketing campaign. For example, electronic messages, printed letters, product samples, coupons, product and other “useful” information may be conveyed to an individual, and other contacts, such as telephone calls, may be initiated.
- An individual may be provided with an “opt-in” invitation, or the individual may be solicited for input, advice or interviews.
- a contact may acknowledge and/or celebrate an individual's expertise, knowledge and/or celebrity, and may invite an individual to participate in an event such as a focus group. Other contacts that create a favorable impression to the individual may also be initiated.
- an additional measurement/feedback system 26 that receives feedback (designated at 28 ) from either or both of worlds 14 , 16 for use in determining marketing efficacy.
- Various known forms of collecting and processing feedback may be used to gauge the effectiveness of a particular marketing encounter and/or campaign. For example, emails capable of triggering a report back to a monitor may be used to detect viewings by connectors, as well as forwarding of the emails to additional recipients. Questionnaires and surveys may also be directed to connectors to gauge effectiveness.
- Additional operations may be desirable in a viral marketing architecture. For example, it may be desirable to convert electronic identities into real identities. In many forums, for example, individuals are identified by email address or pseudonym, and it is often only these identifiers that will be available during data mining and analysis. Translation, e.g., via directory lookup, examination of message content, examination of user profiles, or working with a third-party host of a message archive may be used to derive a real identity to which encounters may be directed. In the alternative, an encounter may be directed to the electronic identity, e.g., by posting or sending a reply directed to a specific individual.
- connectors may be offered surveys, or their messages may be analyzed manually or automatically to determine their usefulness.
- communication assessment may be used to identify individuals for market research projects, e.g., to select individuals for placement in a focus group, to select individuals from which to solicit input on product/service improvements, etc.
- Communication assessment may also be used to select individuals having high listener strengths, e.g., to identify individuals that might be more receptive to listening to a particular proposal, message or encounter.
- FIG. 2 next illustrates an exemplary hardware and software environment for an apparatus 30 suitable for implementing communication assessment consistent with the invention.
- apparatus 30 may represent practically any type of computer, computer system or other programmable electronic device, including a client computer, a desktop computer, a server computer, a portable computer, a handheld computer, an embedded controller, etc.
- Apparatus 30 will hereinafter also be referred to as a “computer”, although it should be appreciated the term “apparatus” may also include other suitable programmable electronic devices consistent with the invention.
- Computer 30 typically includes at least one processor 31 coupled to a memory 32 .
- Processor 31 may represent one or more processors (e.g., microprocessors), and memory 32 may represent the random access memory (RAM) devices comprising the main storage of computer 30 , as well as any supplemental levels of memory, e.g., cache memories, non-volatile or backup memories (e.g., programmable or flash memories), read-only memories, etc.
- RAM random access memory
- memory 32 may be considered to include memory storage physically located elsewhere in computer 30 , e.g., any cache memory in a processor 31 , as well as any storage capacity used as a virtual memory, e.g., as stored on a mass storage device 35 or on another computer coupled to computer 30 via a network.
- Computer 30 also typically receives a number of inputs and outputs for communicating information externally.
- computer 30 typically includes one or more user input devices 33 (e.g., a keyboard, a mouse, a trackball, a joystick, a touchpad, and/or a microphone, among others) and a display 34 (e.g., a CRT monitor, an LCD display panel, and/or a speaker, among others).
- user input may be received via another computer interfaced with computer 30 , e.g., a workstation.
- computer 30 may also include one or more mass storage devices 35 , e.g., a floppy or other removable disk drive, a hard disk drive, a direct access storage device (DASD), an optical drive (e.g., a CD drive, a DVD drive, etc.), and/or a tape drive, among others.
- mass storage devices 35 e.g., a floppy or other removable disk drive, a hard disk drive, a direct access storage device (DASD), an optical drive (e.g., a CD drive, a DVD drive, etc.), and/or a tape drive, among others.
- computer 30 may include an interface with one or more networks (e.g., a LAN, a WAN, a wireless network, and/or the Internet, among others) via one or more network adaptors 36 to permit the communication of information with other computers coupled to the network.
- network adaptors 36 e.g., a LAN, a WAN, a wireless network, and/or the Internet, among others
- Computer 30 operates under the control of an operating system 38 , and executes or otherwise relies upon various computer software applications, components, programs, objects, modules, data structures, etc. (e.g., database management system (DBMS) 40 , local message archive 42 , study 44 , data mining engine 46 , and quantitative analysis engine 48 , among others).
- various applications, components, programs, objects, modules, etc. may also execute on one or more processors in another computer coupled to computer 30 via a network 36 , e.g., in a distributed or client-server computing environment, whereby the processing required to implement the functions of a computer program may be allocated to multiple computers over a network.
- DBMS database management system
- the functions of harvesting messages, managing message archives and study data, and performing quantitative analysis may be performed by different computers interfaced together over a network.
- routines executed to implement the embodiments of the invention whether implemented as part of an operating system or a specific application, component, program, object, module or sequence of instructions will be referred to herein as “computer programs”, or simply “programs”.
- the computer programs typically comprise one or more instructions that are resident at various times in various memory and storage devices in a computer, and that, when read and executed by one or more processors in a computer, cause that computer to perform the steps necessary to execute steps or elements embodying the various aspects of the invention.
- signal bearing media include but are not limited to recordable type media such as volatile and non-volatile memory devices, floppy and other removable disks, hard disk drives, magnetic tape, optical disks (e.g., CD-ROM's, DVD's, etc.), among others, and transmission type media such as digital and analog communication links.
- program may actually incorporate program code executing on multiple computers, as well as program code that may be considered to be executing as multiple computer applications.
- program code performing certain aspects of the illustrated embodiment within a quantitative analysis engine may nonetheless still be considered to be one program within the context of the invention, irrespective of the fact that such program code may be executing in different applications and on different computers.
- FIGS. 1 and 2 are not intended to limit the present invention. Indeed, those skilled in the art will recognize that other alternative hardware and/or software environments may be used without departing from the scope of the invention.
- FIG. 3 presents an overview of the principal steps utilized in an exemplary implementation of a communication assessment system consistent with the invention.
- public message archives (represented at 50 ) are harvested by data mining engine 46 of apparatus 30 ( FIG. 2 ) in a harvest step 52 , resulting in the population of local message archive 42 in database management system 40 .
- the public message archive can represent practically any electronically-accessible message repository or collection of message repositories, e.g., news archives, forums, mailing list (listserv) servers, instant messaging archives, chat archives, telephone record archives, etc., and may be accessible via the Internet or another communication network.
- Harvesting is generally used to pare down the available search space to focus on messages directed to a particular domain of interest, e.g., “water purification”, or “teens' interest in electronic games.”
- Harvesting typically incorporates identifying electronic message repositories that potentially contain messages relevant to the domain of interest.
- specific news archives or forums that potentially pertain to a particular domain of interest are utilized as the principal search space from which harvesting occurs.
- Harvesting may be performed in an automated fashion using any of a number of known harvesting techniques, e.g., using various commercially-available data mining engines such as Content Exchange available from Liason Technology.
- data records are harvested (retrieved and saved) from the search space and organized in the local message archive, optionally with the assistance of automated agents.
- the data harvested may contain as many of the fields describing a forum posting as are desired or available, e.g., date posted, time posted, author ID, message forum where posted, message text, thread ID, response message ID, etc.
- web links (URL's) to the forum posts may also be harvested.
- a local archive may be maintained in any of a number of database architectures.
- a local archive may be maintained in any of a number of commercially available database management systems, e.g., Microsoft Access, Microsoft SQL server, Oracle, etc.
- Harvesting to a local archive is often desirable for performance reasons, given that harvesting can be substantially automated, and that access from a local archive during selection and model building is often more reliable and faster than accessing directly from public archives. Moreover, segregating harvesting and study selection permits study variations to be reframed relatively easily to reformulate results to account for different input parameters. In the alternative, however, a local message archive may not be used and/or harvesting may not performed in some instances.
- a select step 54 is typically performed to further reduce the search space from which relevant electronic messages are obtained, and to create therefrom the framework for a study 44 .
- keywords may be selected to narrow the topic of interest further, as well as to limit messages to certain date ranges, forums, etc., and to select out certain exceptional data if necessary.
- various study parameters are selected, including parameters for use in quantitative analysis, as well as definitional parameters such as what is a conversation, who is a talker and who is a listener.
- individual individuals, forums, threads, etc. may also be excluded from a study (e.g., if a certain individual or forum is known to have little relevance to a particular study).
- the definitional parameters such as what is a conversation (a conversation criterion), who is a talker (a talker criterion) and who is a listener (a listener criterion) will typically vary in different applications, and may be selectable and controllable to vary study results within a particular application.
- a conversation in terms of written communications, may be defined even in the case of one-way messages such as e-mails.
- a conversation includes a message, an author who created the message, and a recipient (who is believed to have read the message).
- a conversation is considered to be between two individuals, so that a message sent to multiple individuals will constitute multiple conversations between the same author and each of the recipients.
- Conversations need not require proof that an intended recipient actually read a message.
- a conversation may require that a recipient reply to or forward a message to confirm receipt of a message, or have posted a message in the same thread. Otherwise, simple membership on a distribution list, or membership in a forum, may be sufficient to identify a recipient of a message posted to the list or forum.
- conversations may be defined by looking at archives of transcripts, and may require additional analysis to detect who was listening whenever a particular individual sent a message, e.g., based upon who replied directly to a statement, or even who was logged on at the time.
- a “talker” can be defined as the author of a message, or only the author of a particular type of message (e.g., the poster of the first message in a thread).
- a talker may be required to have some posting activity above a certain threshold (e.g., more than two posts).
- Listeners may simply be selected from the remaining individuals represented in an archive, or may be further restricted, e.g., to those that participated in one or more forums with a certain degree of activity beyond simple membership. Listeners, for example, may be required to have responded to a message or participated in a thread.
- FIGS. 4 and 5 illustrate different manners of defining talker and listener criteria in the exemplary implementation within the context of forum archives.
- a search space may include a plurality of forums 80 , each of which including a plurality of posts 82 representing the electronic messages being analyzed.
- Certain posts 82 may be organized into threads (e.g., thread 84 ) whereby a chain of posts is defined with later posts in the chain consisting of replies to the original post or other earlier posts in the chain (in some archives, tree-like structures may also be defined to represent sub-threads that depart when a post is in reply to an intermediate post in the thread).
- FIG. 4 illustrates a relatively restrictive definition, where a talker is an author of a first, or initial, post in a thread.
- a listener in a restrictive definition might only be those individuals who can be confirmed to have read an author's post.
- listeners may be restricted to those who post replies to a particular post, or participate in a particular thread.
- an individual 86 who initiates thread 82 may be designated a talker, while the individuals who reply to the initial post (designated at 88 and 90 ) may be designated listeners.
- two conversations are defined: one between individuals 86 and 88 , and one between individuals 86 and 90 .
- a comparatively looser set of criteria would be to distinguish talkers and listeners by virtue of the number of posts made by individuals to the search space.
- a certain threshold e.g., 3 posts or more
- talkers and/or listeners may be used consistent with the invention. For example, listeners could be designated by virtue of merely participating in a thread, specifically replying to a particular post, or just via membership in a forum. Moreover, in some instances it may be desirable to incorporate multiple factors, as well as scale such factors to effectively assign different weights to different types of activities. Also, for certain individuals within a domain of interest that do not meet minimum thresholds for talking and/or listening, such individuals may be excluded from analysis entirely.
- selection of messages from archive 42 may be performed in select step 54 in a number of manners. For example, one versed in SQL or other database languages may build a study 44 through the generation of appropriate database statements and instructions.
- a user interface e.g., a graphical user interface, may be used to permit a user lacking specific knowledge of database languages to perform similar operations.
- a study may also specify a particular study domain, representing the narrowing of the harvest domain by a desired selection criterion.
- a study domain may be constrained, for example, by dates, forums, keywords/content, ad hoc corrections/exclusions (e.g., irrelevant individuals or forums), etc.
- affective elements e.g., representing the strength of the language used in messages, the presence of certain types of persuasive words, usages of particular semantic structures, etc., may also be considered, as may be the overall tone of messages (e.g., positive or negative).
- a study 44 once selected, will include a database of “conversations”, with each identifying, in the least, an author and a recipient.
- the conversation database may also be processed and organized to generate a conversation table that identifies in each entry an author, a recipient, and a conversation count, optionally weighted, representing the number of conversations found between the associated author and recipient.
- quantitative analysis engine 48 is initiated to perform a build model step 56 to generate a mathematical model that expresses the dual propositions of strong talkers talking frequently to strong listeners, and strong listeners listening frequently to strong talkers.
- quantitative analysis engine 48 is implemented using MatLab 6, which is available from The Mathworks, Inc. of Natick, Mass. Other technical computing tools may be used in the alternative.
- results are typically provided to a user via a report step 60 , which may include any desired presentation of results, including text, tables, figures, graphs, charts, plots, etc. As will become more apparent below, any number of suitable user interfaces may be used to present the results to the user.
- FIGS. 6A and 6B next illustrate an exemplary database schema for representing the electronic messages and individuals within a local message archive consistent with the invention.
- Tables 110 include data from the harvesting operation, and comprise the local message archive 42 of FIG. 2 .
- Other tables in the diagram may be created by scripts, as described below, or may be generated in other manners known in the art.
- Table 120 collects data from tables 110 and screens them for study criteria, including date ranges, for example.
- Table 130 constitutes a study (item 44 in FIG. 2 ), based on a loose definition of a conversation.
- Table 132 extracts all the unique talkers from table 130 .
- Table 134 extracts all the unique listeners from table 130 .
- table 140 constitutes another study, this one based on a tight definition of a conversation.
- Table 142 extracts the unique talkers from table 140
- table 144 extracts the unique listeners from table 140 .
- scripts may be used to create, modify, or delete tables in database 40 of FIG. 3 .
- An exemplary script as shown below in Table I:
- the actual messages may be collected within an archive, or that links to such messages may be collected so that a user can retrieve the messages from the original, public archive, when reviewing the results.
- FIGS. 7–9 illustrate in greater detail the quantitative analysis operations utilized in the illustrated implementation to derive results from a study consistent with the invention.
- the mathematical bases behind the operations performed during quantitative analysis are first explained hereinafter, assuming a message archive based upon forums (discussion groups).
- D be a domain of interest, including
- each value W rs may represent a “conversation count” between users r and s.
- some other value referred to as an activity indicator
- activity indicators such as normalized conversation counts, rate of conversations (number per unit of time), regularity of conversations (variance of time between messages), average length of text in messages, etc., may be used.
- the convergence criterion can be represented as
- the vector u is normalized at every iteration, but the vector v is not.
- v 1
- each user in T may be considered to be a strong talker in D since he or she talks to a large number of strong listeners.
- each user in L may be considered to be a strong listener in D since he or she listens to strong talkers.
- the best connectors may be selected from those individuals having the highest talking strengths, and possibly also having the highest listening strengths as well (e.g., a higher listening strength may be used to favor one individual having the same talking strength as another individual).
- an additional concept of “wearout” may be factored into resolving communication strengths.
- diminishing returns are obtained through repeated viewings of an advertisement.
- the incremental strength may diminish for greater numbers of messages to the same listener.
- the diminishing returns of repeated messages may be factored into the determinations of relative communication strengths.
- the half-life h represents one implementation of a “wearout factor”, which is used to mathematically represent the diminishing effect of messages. It will be appreciated that other forms of wearout factors may be used to transform conversation counts consistent with the invention.
- h low typically should be selected to be large enough that g(w max ) doesn't “vanish.”
- h high maybe chosen arbitrarily high (representing “no wearout”), or to assure some nominal wearout level ⁇ u .
- ⁇ l and ⁇ u representing upper and lower attenuations (for the maximum message frequency).
- h low min ⁇ ⁇ h
- h high maxh
- routine 200 begins in block 202 by setting the parameters for use in building the model, including the number of iterations for the approximation algorithm used to solve the eigenvalue problem (the iteration limit), the convergence tolerance ( ⁇ , used in equation (7a)), a study identifier, and details regarding the desired report values and plots to be generated by engine 48 (plots and reports may be generated by either of engine 48 or DBMS 40 depending upon the relative capabilities of each).
- Additional variables e.g., the averaging factor ( ⁇ , used in equation (7)) and the upper and lower attenuations ( ⁇ l and ⁇ u , used in equations (12) and (13)) may also be specified in the study. Any of the above variables may also be fixed and unalterable by a user.
- talker-talker conversation counts are retrieved into a working matrix.
- the conversation table stored in the study is accessed to retrieve into a talker-talker area of a working matrix the conversation counts between individuals defined to be “talkers” in the study, i.e., those individuals that meet the defined talker criterion. Since any given individual may both receive and send messages, it will be appreciated that two different conversation counts will typically be associated with any given pair of individuals, one reflecting conversations where the first individual is the author and the second is the recipient, and one reflecting conversations where the first is the recipient and the second is the author.
- talker-listener conversation counts are retrieved into the same working matrix, this time to store the conversation counts from the same set of talkers to other individuals that meet the listener criterion.
- the diagonal entries in the talker-talker area of the working matrix are nulled (zeroed out) given that any non-zero diagonal entries represent conversations where the same individual is both author and recipient for a conversation under a given conversation criterion (e.g., which might occur if a listener of a message is considered to be any member of a forum, and where the author of the message is a member of the forum), which is not generally useful information for a study of this type.
- FIG. 8 illustrates an exemplary working matrix 220 capable of being generated by the operations described above in connection with blocks 208 .
- the authors of conversations are arranged in the rows of the matrix, while the recipients are arranged in the columns of the matrix. Each entry in the matrix then stores a conversation count for a given pair of author and recipient.
- talker-talker and talker-listener areas 222 , 224 are defined, representing the areas in which the author is a talker, and the recipient is respectively a talker or listener).
- Listener-talker and listener-listener areas of the matrix (collectively represented at 226 ) are either not used, and nulled out, or are simply omitted from the dimensions of the matrix.
- entries 228 the diagonal entries in the talker-talker area 222 are nulled out by virtue of block 208 ( FIG. 7 ).
- a “sparse” matrix is typically generated in block 210 , which assists in minimizing the calculation overhead for engine 48 , and thus improving performance.
- the use and generation of sparse matrices are well known in the art.
- reach and frequency wearout factors are optionally calculated as described above, and routine 200 is complete. Where wearout factors are not used, blocks 212 and 214 are omitted.
- a solve routine 240 is illustrated for implementing solve step 58 ( FIG. 3 ).
- multiple cycles may be performed depending upon the number of wearout factors specified, as represented by FOR loop block 242 .
- control passes to block 244 to calculate a wearout factor transform table, which may be implemented, for example, as a lookup table with an entry for each conversation count between one and the maximum conversation count seen in the working matrix.
- a wearout factor transform table which may be implemented, for example, as a lookup table with an entry for each conversation count between one and the maximum conversation count seen in the working matrix.
- the associated conversation count is scaled using equation (9) above and stored in the entry.
- the working matrix is transformed using the transform table, essentially by replacing every conversation count in the working matrix by the scaled value stored in the transform table entry for that conversation count.
- no transform table may be used, whereby the calculation representing equation (9) may be performed for each matrix entry, typically with lower performance. Also, where wearout factors are not used, blocks 242 , 244 and 246 would not be performed.
- the eigenvalue problem discussed above is solved by the quantitative analysis engine, generally by using the aforementioned parameters set in block 202 of FIG. 7 , choosing an initial value for u of all one's, normalizing u, calculating v (equation (4) above), calculating a revised value for u (equation (7) above), comparing the change in u to ⁇ (equation (7a) above), and repeating until either the change falls below the convergence criterion or the iteration limit has been reached.
- control returns to block 242 to process the working matrix using additional wearout factors.
- control passes to block 252 to generate any reports, plots, graphs, etc., desired from the quantitative analysis engine. Routine 240 is then complete. Alternatively, reports and graphs may be generated after solution for each wearout factor.
- reports, plots, graphs, and the like may be generated.
- reports regarding comparative communication strengths i.e., talking strengths, listening strengths, or composite values representing combined talking and listening strengths
- Talkers/listeners can be listed by forum, and links can be provided within reports to particular forums, user profiles, threads, posts, etc.
- the links may be made to the local and/or public message archives, as well as to other sources of author (individual) information such as home pages, user profile documents, directory listings, search query results, etc.
- various graphs representing communication strengths may be generated and presented to a user.
- FIG. 10 illustrates an exemplary talker strength diagram capable of being generated by a quantitative analysis engine to display the relative talking strength of various talkers 302 in a horizontal bar chart form.
- a threshold represented by vertical line 306
- a predetermined percentage of talkers may be designated as connectors; e.g., by selecting the top x % of all talkers based upon talking strength.
- similar diagrams may be generated for listening strength, as well as composite communication strengths that incorporate both talking and listening strength.
- FIG. 11 illustrates an exemplary HTML-compatible browser 310 from which study results may be presented to a user.
- results are displayed to a user via HTML-compatible documents containing hypertext links to various report pages, charts, and further information about a particular study.
- Such documents may be generated by a database management system, by a quantitative analysis engine, or by a separate application interfaced therewith.
- a study results document may include a link 312 to another document that identifies the study parameters, so that a user can view the conditions under which a study was conducted.
- FIG. 11 illustrates a format in which various reporting options are presented in multiple columns, one for each wearout factor scenario utilized in computing study results.
- wearout factors represented by half-life values of 2 and 32, are used to present results based upon reach valued and frequency valued scenarios.
- Links 314 and 316 provide more detailed information on each individual identified as a connector in each scenario, typically including a strength rating and links to the various messages posted by such connectors.
- Links 318 and 320 provide the mapping of e-mail names or addresses for any connectors, including pseudonyms and optionally any identification information obtained for the connectors.
- Links 322 and 324 permit a user to retrieve links to the various messages posted by such connectors, often arranged by forum or other organization.
- Links 326 and 328 provide access to all the messages posted by all talker individuals, including connectors and non-connectors.
- Link 330 provides access to all messages in the study, in the local or public messaging archives organized by forum and poster, while link 332 provides access to the various posters that authored messages in the archives.
- members may be associated with post counts representing the number of messages posted by those members.
- Additional information e.g., the descriptive information shown at 334 , may also be presented to a user.
- additional links such as link 336 , additional features regarding concepts represented in a result, including a tutorial or overview on the various concepts utilized in the study, may be provided in separate documents.
- links such as link 338 may be provided for access to various plots and graphs generated in connection with the study (here, a bubbleplot diagram such as shown in FIG. 12 ).
- FIGS. 12 and 13 illustrate a bubbleplot diagram 340 and a grouped talker strength diagram 342 capable of being generated in connection with the presentation of study results.
- the bubbleplot diagram 340 may be useful in analyzing the communication strength of various talkers based upon different wearout factor influences.
- Each bubble plotted on the diagram represents a talking strength under two scenarios, a frequency scenario and a reach scenario, each of which is used for one of the orthogonal axes in the diagram.
- talkers are grouped together by profile, such that all the talkers that meet this particular set of talker strengths (or are within a certain range) are grouped together, such that the size of the bubble plotted at a particular point on the diagram represents a relative size of the membership meeting the associated profile.
- the diagram in FIG. 13 groups similar individuals together into groups, and utilizes separate vertical bar graphs to represent (1) the normalized index in terms of talking strength for each group (graph 344 ), and (2) the relative membership of individuals within each group (graphs 346 ). In other studies, it may also be desirable to incorporate within bar graph 344 the relative indexes under each wearout factor scenario, as well.
- individuals may be selected via manual selection or via an automated mechanism, typically based upon a selection criterion based at least in part on communication strength.
- marketing activities may then be initiated using those individuals selected from the study based upon their determined communication strength, e.g., to direct a viral marketing encounter to such individuals, to further screen such individuals, to invite those individuals to participate in a focus group or market research project and/or to monitor the effectiveness of encounters with such individuals.
- a select step was performed to shape the domain into a more compact set for analysis:
- the matrix of message activity (# of messages) from each Talker to each person was constructed.
- the size of the matrix was 193 ⁇ 731, and was processed by Matlab 6 to solve the eigenvalue problem in the manner described above.
- the results generated included a set of relative strength weights assigned to each Talker, indexed to maximum. Listener strengths were also produced.
- the talkers with high strength ratings e.g. talking strength indexes of at least 15
- connectors were designated “connectors”, and from this selection criterion, 25 connectors were identified.
Abstract
Description
TABLE I |
Exemplary SQL Script |
DROP TABLE [Response_Traffic_table] |
GO |
CREATE TABLE [Response_Traffic_table] ( |
[Talker] [varchar] (50) |
COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL, |
[club_name] [varchar] (100) |
COLLATE SQL_Latin1_General_CPI_CI_AS NULL, |
[message_date] [datetime]NULL, |
[message_id] [int]NOT NULL, |
[reply_id] [int]NULL, |
[Listener] [varchar](50) |
COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL, |
[LDate] [datetime]NULL, |
[message_url] [varchar](200) |
COLLATE SQL_Latin1_General_CP1_CI_AS NOT NULL |
) ON [PRIMARY] |
GO |
INSERT INTO [Response_Traffic_table] |
SELECT DISTINCT |
TOP 100 PERCENT dbo.merge_table.member_id AS Talker, |
dbo.merge_table.club_name, dbo.merge_table.message_date, |
dbo.merge_table.message_id, dbo.merge table.reply_id, |
merge_table_1.member_id AS Listener, |
merge_table_1.message_date AS LDate, |
dbo.merge_table.message_url |
FROM dbo.merge_table INNER JOIN dbo.merge_table merge_table_1 |
dbo.merge_table.reply_id = merge_table_1.message_id AND |
dbo.merge_table.club_name = merge_table_1.club_name |
WHERE (dbo.merge_table.message_id IS NOT NULL) AND |
(dbo.merge_table.message_date > |
CONVERT(DATETIME, ‘2000-10-10 00:00:00’, 102)) AND |
(dbo.merge_table.message_date < CONVERT(DATETIME, |
‘2001-04-10 00:00:00’, 102)) |
AND (dbo.merge_table.member_id IS NOT NULL) |
ORDER BY dbo.merge_table.member_id |
GO |
The above exemplary script deletes an existing table such as table 140 in
-
- a set of forums (discussion groups),
- a set of users (individuals), and
- a set of discussion threads.
-
- Ui be user i
- Fj be forum j
- Tjk be thread k in Fj
- Wijk t=talk activity for Ui with respect to Tjk
- Wijk l=listen activity for Ui with respect to Tjk
Wijk t=1 if Ui submitted a message to Tjk, 0 otherwise (1a)
or
Wijk t=1 if Ui submitted initial message of Tjk, 0 otherwise (1b)
or
Wijk t=percentage of messages in Tjk submitted by Ui (1c)
or
Wijk t=number of messages in Tjk submitted by Ui (1d)
Wijk l=1 if Ui is a member of Fj, 0 otherwise (2a)
or
Wijk l=1 if Ui submitted a message in Tjk, 0 otherwise (2b)
or
Wijk l=percentage of messages in Tjk submitted by Ui (2c)
or
Wijk l=number of messages in Tjk submitted by Ui (2d)
-
- W=matrix of values Wrs representing messaging activity
- u=vector of overall talking strength, such that ui is for user Ui
- v=vector of overall listening strength, such that vi is for user Ui
v=Wu
u=WTv (4)
u=WTWu (5)
u=1 (6)
and iteratively apply equation (4) until the values of u and v stabilize. For example, at each iteration, the root mean square of change in either/both vectors can be checked against a tolerance for stopping. Also, at each iteration, the u vector may be revised to a weighted average of an incumbent value and a new value, such as:
u new=(1−α)u old +αW T v (7)
-
- T be the subset of users such that uj is large (e.g. top x %),
-
- L be the subset of users such that vj is large (e.g. top x %)
f(n)=p n−1 (8)
and the cumulative strength of n messages is
f(h)=p h−1=0.5
so
W′=F(W) (11)
such that
w′ ij =F(w ij)
-
- hlow, which represents fast wearout, and rewards reach,
-
- hhigh, which represents slow wearout, and rewards frequency.
w max=max(w ij).
-
- Forums with fewer than 10 messages were dropped, leaving 58 forums to study. Only the 2553 messages in those forums and their 815 authors were retained for analysis. These authors were treated as potentially both talkers and listeners as explained below.
- Among several alternative definitions, the following was chosen to constitute “a message from one person to another”: If a person posted a message to a forum then everyone else in the forum heard it. Only members who actively posted to the forum and on the subject defined in the data-mining were included as listeners, to exclude lurkers or inactive members.
- At this point the 731 persons who authored messages were classified. Among them, 193 were designated as “Talkers.” They were considered a Talker if they had more than 1 message to at least some person. The remaining 538 individuals were designated “Listeners.”
Claims (18)
v=Wu
u=WTv
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/897,577 US7143054B2 (en) | 2001-07-02 | 2001-07-02 | Assessment of communication strengths of individuals from electronic messages |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/897,577 US7143054B2 (en) | 2001-07-02 | 2001-07-02 | Assessment of communication strengths of individuals from electronic messages |
Publications (2)
Publication Number | Publication Date |
---|---|
US20030055711A1 US20030055711A1 (en) | 2003-03-20 |
US7143054B2 true US7143054B2 (en) | 2006-11-28 |
Family
ID=25408071
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/897,577 Expired - Lifetime US7143054B2 (en) | 2001-07-02 | 2001-07-02 | Assessment of communication strengths of individuals from electronic messages |
Country Status (1)
Country | Link |
---|---|
US (1) | US7143054B2 (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050154556A1 (en) * | 2004-01-13 | 2005-07-14 | Keller Edward B. | System and method of identifying individuals of influence |
US20060036456A1 (en) * | 2004-08-12 | 2006-02-16 | Yuh-Cherng Wu | Virtual community generation |
US20060259473A1 (en) * | 2005-05-13 | 2006-11-16 | Microsoft Corporation | System and method for utilizing the content of an online conversation to select advertising content and/or other relevant information for display |
US20080235670A1 (en) * | 2005-05-05 | 2008-09-25 | International Business Machines Corporation | Method and Apparatus for Creation of an Interface for Constructing Conversational Policies |
US20100205037A1 (en) * | 2009-02-10 | 2010-08-12 | Jan Besehanic | Methods and apparatus to associate demographic and geographic information with influential consumer relationships |
US20100217720A1 (en) * | 2009-02-20 | 2010-08-26 | Microsoft Corporation | Identifying users for effective propagation of content |
US20100217671A1 (en) * | 2009-02-23 | 2010-08-26 | Hyung-Dong Lee | Method and apparatus for extracting advertisement keywords in association with situations of video scenes |
US20110131144A1 (en) * | 2009-11-30 | 2011-06-02 | International Business Machines Corporation | Social analysis in multi-participant meetings |
US8332477B1 (en) * | 2011-08-25 | 2012-12-11 | Google Inc. | Presenting related communications |
US20140095418A1 (en) * | 2009-03-23 | 2014-04-03 | Andrew J. Scholand | Framework for quantitative analysis of a communication corpus |
US8880629B1 (en) | 2012-06-18 | 2014-11-04 | Kabam, Inc. | Dynamically providing system communications tailored to individual users responsive to trigger events in virtual spaces |
US9098387B1 (en) | 2013-02-22 | 2015-08-04 | Kabam, Inc. | System and method for providing a customized user experience based on a spend frequency of a user |
US10320927B2 (en) * | 2016-10-20 | 2019-06-11 | Facebook, Inc. | Systems and methods for providing personalized content |
Families Citing this family (42)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7519542B1 (en) * | 2001-08-14 | 2009-04-14 | Versata Development Group, Inc. | System and method for modeling and applying a people network representation |
US20120060102A1 (en) | 2002-08-07 | 2012-03-08 | Joseph Shohfi | System and method for visual communication between buyers and sellers |
US7426693B2 (en) * | 2002-08-07 | 2008-09-16 | Global Apparel Network, Inc. | Visual communications tool |
JP5179757B2 (en) * | 2004-02-17 | 2013-04-10 | トランセプト ファーマシューティカルズ, インコーポレイティド | Composition for delivering a hypnotic agent through the oral mucosa and method of use thereof |
US20050204004A1 (en) * | 2004-03-11 | 2005-09-15 | International Business Machines Corporation | Method and apparatus for managing status in a messaging program |
US9819624B2 (en) | 2004-03-31 | 2017-11-14 | Google Inc. | Displaying conversations in a conversation-based email system |
US7269621B2 (en) | 2004-03-31 | 2007-09-11 | Google Inc. | Method system and graphical user interface for dynamically updating transmission characteristics in a web mail reply |
US7814155B2 (en) | 2004-03-31 | 2010-10-12 | Google Inc. | Email conversation management system |
US7912904B2 (en) * | 2004-03-31 | 2011-03-22 | Google Inc. | Email system with conversation-centric user interface |
US7437382B2 (en) * | 2004-05-14 | 2008-10-14 | Microsoft Corporation | Method and system for ranking messages of discussion threads |
US7979501B1 (en) * | 2004-08-06 | 2011-07-12 | Google Inc. | Enhanced message display |
FI20041323A (en) * | 2004-10-12 | 2006-04-13 | Xtract Oy | Analyzer, system, and method for determining the desired user population |
US9002725B1 (en) | 2005-04-20 | 2015-04-07 | Google Inc. | System and method for targeting information based on message content |
US20060253524A1 (en) * | 2005-05-05 | 2006-11-09 | Foreman Paul E | Representations of conversational policies |
US7949714B1 (en) * | 2005-12-05 | 2011-05-24 | Google Inc. | System and method for targeting advertisements or other information using user geographical information |
US8601004B1 (en) | 2005-12-06 | 2013-12-03 | Google Inc. | System and method for targeting information items based on popularities of the information items |
US20090119173A1 (en) * | 2006-02-28 | 2009-05-07 | Buzzlogic, Inc. | System and Method For Advertisement Targeting of Conversations in Social Media |
US20070214097A1 (en) * | 2006-02-28 | 2007-09-13 | Todd Parsons | Social analytics system and method for analyzing conversations in social media |
US20080046317A1 (en) * | 2006-08-21 | 2008-02-21 | The Procter & Gamble Company | Systems and methods for predicting the efficacy of a marketing message |
US9213554B2 (en) * | 2006-12-04 | 2015-12-15 | Sap Se | Generic chart interface |
US20080140506A1 (en) * | 2006-12-08 | 2008-06-12 | The Procter & Gamble Corporation | Systems and methods for the identification, recruitment, and enrollment of influential members of social groups |
US8484083B2 (en) * | 2007-02-01 | 2013-07-09 | Sri International | Method and apparatus for targeting messages to users in a social network |
US20080215607A1 (en) * | 2007-03-02 | 2008-09-04 | Umbria, Inc. | Tribe or group-based analysis of social media including generating intelligence from a tribe's weblogs or blogs |
GB2448136A (en) * | 2007-04-02 | 2008-10-08 | Motorola Inc | A content item distribution system |
US20080306899A1 (en) * | 2007-06-07 | 2008-12-11 | Gregory Michelle L | Methods, apparatus, and computer-readable media for analyzing conversational-type data |
JP2009009355A (en) * | 2007-06-28 | 2009-01-15 | Hitachi Ltd | Organization communication visualization system |
GB2457720A (en) * | 2008-02-23 | 2009-08-26 | Philip Thomas Rumsby | Method for laser processing on the opposite sides of thin transparent substrates |
US9002820B2 (en) * | 2008-06-05 | 2015-04-07 | Gary Stephen Shuster | Forum search with time-dependent activity weighting |
US20100161727A1 (en) * | 2008-12-19 | 2010-06-24 | Cisco Technology, Inc. | System and Method for Accelerating a Wide Area Notification |
US8041378B2 (en) | 2008-12-19 | 2011-10-18 | Cisco Technology, Inc. | System and method for providing channel configurations in a communications environment |
US8655792B1 (en) * | 2009-03-27 | 2014-02-18 | Symantec Corporation | Deriving the content of a social network private site based on friend analysis |
US8495142B2 (en) | 2010-03-11 | 2013-07-23 | Cisco Technology, Inc. | System and method for providing data channel management in a network environment |
JP5412539B2 (en) * | 2011-06-07 | 2014-02-12 | 東芝テック株式会社 | Merchandise sales data processing apparatus and control program thereof |
US9009142B2 (en) | 2011-07-27 | 2015-04-14 | Google Inc. | Index entries configured to support both conversation and message based searching |
US9443518B1 (en) | 2011-08-31 | 2016-09-13 | Google Inc. | Text transcript generation from a communication session |
US9135291B2 (en) * | 2011-12-14 | 2015-09-15 | Megathread, Ltd. | System and method for determining similarities between online entities |
US8612211B1 (en) | 2012-09-10 | 2013-12-17 | Google Inc. | Speech recognition and summarization |
US20140351713A1 (en) * | 2013-05-23 | 2014-11-27 | Research In Motion Limited | Method and electronic device for updating communication representations based on activity level |
US20150073777A1 (en) * | 2013-09-06 | 2015-03-12 | Brian Assam | System and method for determining semantics and the probable meaning of words |
US9697198B2 (en) * | 2015-10-05 | 2017-07-04 | International Business Machines Corporation | Guiding a conversation based on cognitive analytics |
AU2017203723A1 (en) * | 2016-06-07 | 2017-12-21 | David Nixon | Meeting management system and process |
US10832275B2 (en) * | 2018-05-25 | 2020-11-10 | At&T Intellectual Property I, L.P. | System for management of requirements-based advertisements |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11161670A (en) * | 1997-12-01 | 1999-06-18 | Ntt Data Corp | Method, device, and system for information filtering |
JPH11328275A (en) * | 1998-05-14 | 1999-11-30 | Nec Corp | Method and system for information filtering |
US6151585A (en) * | 1998-04-24 | 2000-11-21 | Microsoft Corporation | Methods and apparatus for determining or inferring influential rumormongers from resource usage data |
EP1065607A2 (en) * | 1999-06-29 | 2001-01-03 | Xerox Corporation | System and method of predicting a user's rating for an item in a collaborative filtering system |
US20020062368A1 (en) * | 2000-10-11 | 2002-05-23 | David Holtzman | System and method for establishing and evaluating cross community identities in electronic forums |
US6414691B1 (en) * | 1999-11-11 | 2002-07-02 | Hitachi, Ltd. | Method and apparatus for displaying graphs |
US20050049908A2 (en) * | 1999-08-31 | 2005-03-03 | Comsort, Inc. | System for Influence Network Marketing |
-
2001
- 2001-07-02 US US09/897,577 patent/US7143054B2/en not_active Expired - Lifetime
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11161670A (en) * | 1997-12-01 | 1999-06-18 | Ntt Data Corp | Method, device, and system for information filtering |
US6151585A (en) * | 1998-04-24 | 2000-11-21 | Microsoft Corporation | Methods and apparatus for determining or inferring influential rumormongers from resource usage data |
JPH11328275A (en) * | 1998-05-14 | 1999-11-30 | Nec Corp | Method and system for information filtering |
EP1065607A2 (en) * | 1999-06-29 | 2001-01-03 | Xerox Corporation | System and method of predicting a user's rating for an item in a collaborative filtering system |
US20050049908A2 (en) * | 1999-08-31 | 2005-03-03 | Comsort, Inc. | System for Influence Network Marketing |
US6414691B1 (en) * | 1999-11-11 | 2002-07-02 | Hitachi, Ltd. | Method and apparatus for displaying graphs |
US20020062368A1 (en) * | 2000-10-11 | 2002-05-23 | David Holtzman | System and method for establishing and evaluating cross community identities in electronic forums |
Non-Patent Citations (18)
Title |
---|
Amento, B., L. Terveen and W. Hill, "Does 'Authority' Mean Quality Predicting Expert Quality Ratings of Web Documents", AT&T Shannon Laboratories, Florham Park, NJ, SIGR, 2000, pp. 296-303. |
Brin, S. and L. Page, "Anatomy of a Large Scale Hypertextual Web Search Engine", Computer Science Department, Stanford University, Stanford CA, 1998, pages 1-17. |
Burson-Marstellar, , The e-fluentials, 2000, pp. 1-11. |
Gibson, D., J. Kleinberg and P. Ragha Van, "Inferring Web Communities from Link Topology", Proceedings of 9<SUP>th </SUP>ACM Conference on Hypertext and Hypermedia, 1998, pp. 225-234. |
Hayes, B., "Graph Theory in Practice: Part I", American Scientist, Jan.-Feb. 2000, vol. 88, No. 1, pp. 9-13. |
Hayes, B., "Graph Theory in Practice: Part II", American Scientist, Mar.-Apr. 2000, vol. 88, No. 2, pp. 104-109. |
Kleinberg, J., "Authoritative Sources in a Hyperlinked Environment", Proceedings on the ACM-SIAM Symposium on Discrete Algorithms, 1998, pp. 668-677. |
Matthews, R., "Six Degrees of Separation", Worldlink, Jan./Feb. 2000, pp. 17-18. |
Newman, M.E.J., "The Structure of Scientific Collaboration Networks", Proceedings of the national Academy of Sciences USA, 2001, vol. 98, No. 1, pp. 404-409. |
Newman, M.E.J., "Who is the Best Connected Scientist" A Study of Scientific Coauthorship Networks, Santa Fe Institute, Santa Fe, NM, pp. 1-32. |
Price, L. and L. F. Feick, "The Market Maven: A Diffuser of Marketplace Information", Journal of Marketing, Jan. 1987, vol. 51, pp. 83-97. |
Smith, M.A. and A.T. Fiore, "Visualization Components for Persistent Conversations", Microsoft Research, Redmond, WA, ACM SIG CHI 2001, 1999, 8 pages. |
Smith, M.A., "Invisible Crowds in Cyberspace: Mapping the Social Structure of the Usenet", Communities in Cyberspace: Perspectives on New Forms of Social Organization, London, Routledge Press, 1999, pp. 1-35. |
Smith, M.A., "Some Social Implications of Ubiquitous Wireless Networks", ACM Mobile Computing and Communications Review, Apr. 2001, vol. 4 No. 2, 9 pages. |
Smith, M.A., S.D. Farhnham and S.M. Drucker, The Social Life of Small Graphical Chat Spaces, Microsoft Research, CHI Letters, Apr. 2000, vol. 2, Issue 1, pp. 462-469. |
Viegas, F.B. and J.S. Donath, "Chat Circles", MIT Media Lab, May 1999, ACM, CHI '99, Pittsburgh, PA, pp. 9-16. |
Watts, D., "Kevin bacon, the Small-World, and Why It All Matters", Santa Fe Institute Bulletin, Fall 1999, vol. 14, No. 2, 6 pages. |
Whittaker, S., L. Terveen, W. Hill and L. Cherny, "The Dynamics of Mass Interaction", ATT Labs-Research, Florham Park, NJ, 1998, 8 pages. |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050154556A1 (en) * | 2004-01-13 | 2005-07-14 | Keller Edward B. | System and method of identifying individuals of influence |
US7653568B2 (en) * | 2004-01-13 | 2010-01-26 | GfK U.S. Holdings, Inc. | System and method of identifying individuals of influence |
US20060036456A1 (en) * | 2004-08-12 | 2006-02-16 | Yuh-Cherng Wu | Virtual community generation |
US7444315B2 (en) * | 2004-08-12 | 2008-10-28 | Sap Aktiengesellschaft | Virtual community generation |
US8266517B2 (en) | 2005-05-05 | 2012-09-11 | International Business Machines Corporation | Creation of an interface for constructing conversational policies |
US20080235670A1 (en) * | 2005-05-05 | 2008-09-25 | International Business Machines Corporation | Method and Apparatus for Creation of an Interface for Constructing Conversational Policies |
US7653627B2 (en) * | 2005-05-13 | 2010-01-26 | Microsoft Corporation | System and method for utilizing the content of an online conversation to select advertising content and/or other relevant information for display |
US20060259473A1 (en) * | 2005-05-13 | 2006-11-16 | Microsoft Corporation | System and method for utilizing the content of an online conversation to select advertising content and/or other relevant information for display |
US20100205037A1 (en) * | 2009-02-10 | 2010-08-12 | Jan Besehanic | Methods and apparatus to associate demographic and geographic information with influential consumer relationships |
US20100217720A1 (en) * | 2009-02-20 | 2010-08-26 | Microsoft Corporation | Identifying users for effective propagation of content |
US20100217671A1 (en) * | 2009-02-23 | 2010-08-26 | Hyung-Dong Lee | Method and apparatus for extracting advertisement keywords in association with situations of video scenes |
US9043860B2 (en) | 2009-02-23 | 2015-05-26 | Samsung Electronics Co., Ltd. | Method and apparatus for extracting advertisement keywords in association with situations of video scenes |
US20140095418A1 (en) * | 2009-03-23 | 2014-04-03 | Andrew J. Scholand | Framework for quantitative analysis of a communication corpus |
US8204759B2 (en) | 2009-11-30 | 2012-06-19 | International Business Machines Corporation | Social analysis in multi-participant meetings |
US8484040B2 (en) | 2009-11-30 | 2013-07-09 | International Business Machines Corporation | Social analysis in multi-participant meetings |
US20110131144A1 (en) * | 2009-11-30 | 2011-06-02 | International Business Machines Corporation | Social analysis in multi-participant meetings |
US8332477B1 (en) * | 2011-08-25 | 2012-12-11 | Google Inc. | Presenting related communications |
US9607078B1 (en) * | 2011-08-25 | 2017-03-28 | Google Inc. | Presenting related communications |
US8880629B1 (en) | 2012-06-18 | 2014-11-04 | Kabam, Inc. | Dynamically providing system communications tailored to individual users responsive to trigger events in virtual spaces |
US10171385B1 (en) | 2012-06-18 | 2019-01-01 | Kabam, Inc. | Dynamically providing system communications in a virtual space tailored to individual users responsive to specific user actions and individual current states |
US10992545B2 (en) | 2012-06-18 | 2021-04-27 | Kabam, Inc. | Dynamically providing system communications in a virtual space tailored to individual users responsive to specific user actions and individual current states |
US9098387B1 (en) | 2013-02-22 | 2015-08-04 | Kabam, Inc. | System and method for providing a customized user experience based on a spend frequency of a user |
US9460443B1 (en) | 2013-02-22 | 2016-10-04 | Kabam, Inc. | System and method for providing a customized user experience based on a spend frequency of a user |
US10058782B2 (en) | 2013-02-22 | 2018-08-28 | Kabam, Inc. | System and method for providing a customized user experience based on a spend frequency of a user |
US10320927B2 (en) * | 2016-10-20 | 2019-06-11 | Facebook, Inc. | Systems and methods for providing personalized content |
US20200145505A1 (en) * | 2016-10-20 | 2020-05-07 | Facebook, Inc. | Systems and methods for providing personalized content |
Also Published As
Publication number | Publication date |
---|---|
US20030055711A1 (en) | 2003-03-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7143054B2 (en) | Assessment of communication strengths of individuals from electronic messages | |
US7197470B1 (en) | System and method for collection analysis of electronic discussion methods | |
US9519936B2 (en) | Method and apparatus for analyzing and applying data related to customer interactions with social media | |
US7827052B2 (en) | Systems and methods for reputation management | |
US20060053156A1 (en) | Systems and methods for developing intelligence from information existing on a network | |
US20020062368A1 (en) | System and method for establishing and evaluating cross community identities in electronic forums | |
US20070124432A1 (en) | System and method for scoring electronic messages | |
US20110125550A1 (en) | Method for determining customer value and potential from social media and other public data sources | |
US8341101B1 (en) | Determining relationships between data items and individuals, and dynamically calculating a metric score based on groups of characteristics | |
US20130339445A1 (en) | Receiving and Correlation of User Choices to Facilitate Recommendations for Peer-to-Peer Connections | |
US7552096B2 (en) | Computer-implemented system and program product for analyzing a collaborative space | |
US20140074612A1 (en) | System and Method for Targeting Information Items Based on Popularities of the Information Items | |
Gittelman et al. | Quota controls in survey research: a test of accuracy and intersource reliability in online samples | |
Ye | Overcoming challenges to conducting online surveys | |
US20170262532A1 (en) | Generating content feeds based on expanded network engagement | |
Marton et al. | A question of quality: The effect of source quality on information seeking by women in IT professions | |
Tyrvainen | Estimating applicability of new mobile content formats to organizational use | |
IBRAHIM et al. | Social Media Monitoring Twitter: Mobile Brand Awareness Using the Selection Sort | |
Diapouli et al. | Behavioural Analytics using Process Mining in On-line Advertising. | |
Juwara et al. | A hybrid covariate microaggregation approach for privacy-preserving logistic regression | |
JP4466279B2 (en) | Information processing apparatus, information processing method, and computer program | |
Rich et al. | New technologies, new approaches to evaluating academic productivity | |
Ansary | The future marketing aspect of Facebook in Bangladesh | |
JP4635454B2 (en) | Evaluation apparatus and method | |
JP4348518B2 (en) | Evaluation apparatus and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: PROCTER & GAMBLE COMPANY, THE, OHIO Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:DOHERTY, MICHAEL EMMETT;REEL/FRAME:012431/0537 Effective date: 20010709 |
|
FEPP | Fee payment procedure |
Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
CC | Certificate of correction | ||
FPAY | Fee payment |
Year of fee payment: 4 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 12TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1553) Year of fee payment: 12 |